SOA Case Study: How R.L. Polk Revved Its Data Engine
Overview
In this paper we will examine the success of R.L. Polk Revved Data Engine.
In 2004 R.L. Polk Company’s CIO made a strategic decision to replace a decades old information system that was the core of its business. Although the system itself was not broken it presented an opportunity for innovation to bring the company forward to for the next 30 years. The CIO was immediately faced the challenge of how to present an expensive internal project with no immediate measureable ROI for the company. Similar projects in the past had failed to gain enough support and subsequently were scaled back. A $20 million project with no foreseeable profit margin was huge risk for Polk.
Approach
The CIO believed that an agile delivery model leveraging newer technology would provide improvements to the legacy IBM batch driven mainframe dinosaur currently in place. Based on the past two failures to convert from the mainframe system the company came up with a different approach to the project by assembling a core team of Polk IT resources and combining them with a group of outside consultants to bring new ideas and agility to the project that it lacked in the past. This was the first building block in the success of the project. The fact the team of consults was able to free up existing Polk resources and allowed them to focus on building the next generation of its engine was what separated them from the failure to migrate from the old mainframe solution in the past. Although it was a bold move to separate the team and create a standalone entry, (RLP Technologies), instead of keeping the team with Polk, the risk of isolating IT from the larger corporate group and not involving the current operations in the project seemed not to impact the project. However I believe the same results could have also been accomplished by building the IT team within the Polk organization.
Additionally some would argue that the cost of so many consultants drove up project cost, however the addition of outside resources infused new ideas and different perspectives of thinking that added to the collective producing a better product. Polk demonstrated the ability to have an open innovative business model, a concept of “You don’t know everything “as discussed in Chesbrough that focuses on leveraging partners to aid in bringing new technology to market. This demonstrated the company’s agility in getting a new data engine to market.
The team’s three high level objectives as state by Polk: “..Referred to as “50/50/100”: The new system need to be 50% more efficient (in other words, cost half as much to operate); deliver data 50% more quickly; and aim for 100% data accuracy”
The second key to success in the project was the CIO’s decision to implement a “service-oriented architecture, or SOA. Polk’s leveraging of Tibco integration software as the backbone provided flexibility in processing data and scalability for hardware configuration not achievable in the mainframe solution. Commercial software product reduced project cost by minimizing the internal development efforts by 50%. This allowed the agility needed to bring the engine to market within 18 months. This would not have been achievable without the open business model approach and incorporating innovation.
Challenges and Success
Here is a comparison of how the project objectives referred to as “50/50/100” rated from the initial charter based on some of the initial challenges faced against the measureable success as determined by the CIO:

Objective | Result | Success Rate Achieved | The new system need to be 50% more efficient | The new data engine removed manual mainframe batch processing that required resource intervention to complete by replacing with a flexible automated process that reduced resource headcount. Additional cost savings were realized in reduced support cost. | Yes | Deliver data 50% more quickly | The process to deliver data to customers was reduced from 3 weeks to hours. | Yes- However no metrics were available for comparison. This was a clear miss in the project baseline. | Aim for 100% data accuracy | The implementation of the data quality analysis software effectively reduced data error rates and cost | Yes | | | |
“We have taken multi-hour processes down to multi-minute processes”~ Kevin Vasconi CIO R.L. Polk
Summary
What lead to the success of this project was the CIO’s non-monetary evaluation presented to the management team that focused on an increase of production, reduction in cost, data accuracy and to improve customer satisfaction in delivery. However the idea that the engine would become a profit center by providing data processing services may not prove successful. One of the risks of using open innovation is that competitors will quickly figure out or improve upon this model and bring it to market.